The long-term goal is to use genetics and epidemiology to identify targets for prevention of colon cancer. The application is on prostaglandin biosynthesis as a promising target, because aspirin prevents some cases of colon cancer. The hypothesis is that genetic variation in prostaglandin production may mimic aspirin effects and shed light on preventive mechanisms. Work will focus on naturally-occurring mutations in people and targeted mutations in mice in a nuclear prostaglandin pathway defined by: cytosolic phospholipase A prostaglandin H synthase 2 (PTGS2/Cox-2), hematopoietic prostaglandin D synthase (H-PGDS), and peroxisome proliferator-activated receptor gamma. The project builds on discovery among 10 percent of African Americans of a PTGS2/Cox-2 mutation (Val5llAla) near the active site of the enzyme. Specific aims are to: (1) conduct case-control analyses on prevalence of colorectal adenomas and cancer in relation to genetic variants in PTGS2/Cox-2 and H-PGDS; (2) assess human interindividual variation in PTGS2/Cox-2 activity in relation to the Va151 1 Ala enzyme variant; (3) develop in vitro expression assays for novel variants identified in H-PGDS; and (4) develop a knockout mouse model of variation in H-Pgds to complement human epidemiologic studies. Three case-control studies will be used to assess effects of genetic variation on colon neoplasms: a Kaiser sigmoidoscopy study of adenomas (1,700 subjects); a Univ. of North Carolina colonoscopy study of adenomas (800 subjects); and African American cancer cases and controls (roughly 400) from the Multiethnic Cohort Study. Results may lead to better understanding of protective mechanisms involving aspirin and nonsteroidal anti-inflammatory drugs.